检索结果-内蒙古大学图书馆

ieee OPEN JOURNAL OF control systems 2024年 3卷 118-127页

作者： Khaledi, Marjan Kiumarsi, Bahare Michigan State Univ Dept Elect & Comp Engn E Lansing MI 48824 USA

This article presents a proactive approach to resolving the conflict between safety and optimality for continuous-time (CT) safety-critical systems with unknown dynamics. The presented method guarantees safety and performance specifications by combining two controllers: a safe controller and an optimal controller. On the one hand, the safe controller is designed using only input and state data measurements and without requiring the state derivative data, which are typically required in data-driven control of CT systems. State derivative measurement is costly, and its approximation introduces noise to the system. On the other hand, the optimal controller is learned using a low-complexity one-shot optimization problem, which again does not rely on prior knowledge of the system dynamics and state derivative data. Compared to existing optimal control learning methods for CT systems, which are typically iterative, a one-shot optimization is considerably more sample-efficient and computationally efficient. The share of optimal and safe controllers in the overall control policy is obtained by solving a computationally efficient optimization problem involving a scalar variable in a data-driven manner. It is shown that the contribution of the safe controller dominates that of the optimal controller when the system's state is close to the safety boundaries, and this domination drops as the system trajectories move away from the safety boundaries. In this case, the optimal controller contributes more to the overall controller. The feasibility and stability of the proposed controller are shown. Finally, the simulation results show the efficacy of the proposed approach.

关键词： Safety Optimal control Optimization control systems System dynamics control design Trajectory control barrier functions (CBFs) data-driven initiative controller optimality safe control unknown systems

来源：评论

学校读者我要写书评

暂无评论

Detecting False data Injection Attacks using Spatial-temporal Graph Neural Network 12

Detecting False Data Injection Attacks using Spatial-tempora...

引用

ieee 12th data driven control and learning systems conference (DDCLS)

作者： Wei, Xingshen Liu, Wei Zhou, Jian Zhou, Xiaoming Zhang, Wenjie Cao, Yongjian Nanjing NARI Informat Commun Technol Co Ltd NARI Grp Corp State Grid Elect Power Res Inst Nanjing Peoples R China State Grid Liaoning Elect Power Supply Co Shenyang Peoples R China

ISBN: (纸本)9798350321050

There are a large number of cyber-attacks in the power system, especially the false data injection attack (FDIA). This attack can bypass the traditional bad data detection mechanism (BDDM), and affect the operation of the power system. In this paper, for the purpose of guaranteeing the reliable operation of the cyber-physical power system (CPPS), a novel FDIA detection model is developed based on spatial-temporal graph neural network (STGNN). The STGNN can extract the temporal features and spatial features of measurement data simultaneously in the CPPS. Specially, the spatial features and the temporal features are extracted by graph neural network (GNN) and recurrent neural network (RNN), respectively. Simulation results based on ieee 14-bus system verify the performance of the proposed method.

关键词： False data injection attack Attack detection Graph convolutional network

来源：评论

学校读者我要写书评

暂无评论

Adaptive robust control of the continuous-time two-input systems with unknown disturbance based on Q-function 12

Adaptive robust control of the continuous-time two-input sys...

引用

ieee 12th data driven control and learning systems conference (DDCLS)

作者： Lv, Yongfeng Cui, Zhengyu Wang, Minlin Taiyuan Univ Technol Coll Elect & Power Engn Taiyuan 030024 Peoples R China Lanzhou Univ Technol Coll Elect & Informat Engn Lanzhou 730050 Peoples R China

ISBN: (纸本)9798350321050

Considering overshoot and chatter of the multi-input system with unknown interference, this paper studies the adaptive robust optimal controls of continuous-time two-input systems with an approximate dynamic programming (ADP) based Q-function scheme. A complex Hamilton-Jacobi-Issacs (HJI) equation is obtained with the two-input system and the zero-game theory, where a value function is constructed. Solving the HJI equation is a challenging task. Thus, an ADP-based Q-function with a neural network is constructed to learn the saddle point of the HJI equation. Simultaneously, an integral reinforcement signal of the critic networks is introduced such that the system drift and input dynamics in the HJI equation are relaxed when studying the saddle-point intractable solution. Then, the adaptive robust optimal actor and worst disturbance are approximated with another three networks. Finally, an F-16 aircraft plant is used to verify the proposed ADP-based Q-function.

关键词： Robust control adaptive control approximate dynamic programming multi-input system

来源：评论

学校读者我要写书评

暂无评论

Transfer Reinforcement learning of Robotic Grasping Training using Neural Networks with Lateral Connections 12

Transfer Reinforcement Learning of Robotic Grasping Training...

引用

ieee 12th data driven control and learning systems conference (DDCLS)

作者： Wang, Wenxiao Wang, Xiaojuan Li, Renqiang Jiang, Haosheng Liu, Ding Ping, Xubin Xidian Univ Sch Elect Mech Engn Xian 710071 Peoples R China

ISBN: (纸本)9798350321050

Reinforcement learning, as an effective framework for solving continuous decision tasks in machine learning, has been widely used in manipulator decision control. However, for manipulator grasping tasks in complex environments, it is difficult for intelligence to improve performance by exploring to obtain high-quality interaction samples. In addition, the training models of reinforcement learning usually lack task generalization and need to be relearned to adapt to task changes. To address these issues, researchers have proposed transfer learning that uses external prior knowledge to help the target task to improve the reinforcement learning process. In this paper, the transfer of the manipulator grasping source task to the grasping target task based on the deep Q-network algorithm is achieved by constructing lateral connections between fully convolutional neural networks using Densenet. Experimental results in the CoppeliaSim simulation environment show that the methods successfully achieve inter-task transfer by constructing lateral connections between fully convolutional neural networks. The validated transfer reinforcement learning approach improves the effectiveness of task training while reducing the complexity of the network due to lateral connections.

关键词： Transfer learning deep reinforcement learning lateral connections manipulator

来源：评论

学校读者我要写书评

暂无评论

Coordinated Voltage Regulation of Microgrid Clusters Based on Deep Reinforcement learning Approach 12

Coordinated Voltage Regulation of Microgrid Clusters Based o...

引用

ieee 12th data driven control and learning systems conference (DDCLS)

作者： Xue, Xiaozhe Ge, Hui Nanjing Normal Univ Sch Elect & Automat Engn Nanjing Peoples R China Nanjing Univ Posts & Telecommun Coll Automat Nanjing Peoples R China Nanjing Univ Posts & Telecommun Coll Artificial Intelligence Nanjing Peoples R China

ISBN: (纸本)9798350321050

With the rapid development of microgrid cluster operation, the problem of voltage regulation in the coordinated operation of multiple microgrids faces practical challenges. Aiming at the problem of voltage regulation of multi-microgrids, this paper firstly establishes an optimization model of coordinated voltage regulation of multiple microgrids considering the coordination of source, grid, load and storage. Since the difficulty of solving the above optimization problem, it is further reformulated as a Markov game. Then, a novel collaborative voltage regulation algorithm based on multi-agent deep reinforcement learning (MADRL) is proposed. In order to improve the scalability of the algorithm, an attention mechanism is introduced into the multi-agent deep reinforcement learning algorithm. The simulation results show that the proposed algorithm can coordinate with multiple microgrids to regulate the voltage to a safe range.

关键词： Multi-microgrids Voltage regulation MADRL Attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Optimal dispatch of an integrated energy system based on deep reinforcement learning considering new energy uncertainty 12

Optimal dispatch of an integrated energy system based on dee...

引用

ieee 12th data driven control and learning systems conference (DDCLS)

作者： Zhou, Yang Jia, Li Zhao, Yilin Zhan, Zhiyong Shanghai Univ Sch Mechatron Engn & Automat Shanghai 200444 Peoples R China

ISBN: (纸本)9798350321050

As the uncertainties of intermittent energy and load in the integrated energy system gradually increase, traditional dispatch methods are limited to fixed physical models and parameter settings that can hardly respond to the random fluctuations in the dynamic system with source-load. In this paper, a deep reinforcement learning-based dynamic dispatch method for the integrated energy system is proposed to address this problem. First, a data-driven deep reinforcement learning model is constructed for the integrated energy system. Through the continuous interaction between the agent and the integrated energy system, the dispatch strategies are learned adaptively to reduce dependence on the physical models. Secondly, the variations of source-load uncertainties are characterized by adding random disturbances. Pivotal aspects such as state spaces, action spaces, reward mechanisms, and the training process of the deep reinforcement learning model are improved according to the characteristics of uncertainties. Then a proximal policy optimization algorithm is used to solve the problem, and the dynamic dispatch decisions of the integrated energy system are realized. Finally, simulation results verify the feasibility and effectiveness of the proposed method over different time scales and in uncertain environments.

关键词： Integrated energy system dynamic dispatch deep reinforcement learning proximal policy optimization

来源：评论

学校读者我要写书评

暂无评论

Fault detection for rolling bearings by multi-sensor information fusion method with adaptive weights 12

Fault detection for rolling bearings by multi-sensor informa...

引用

ieee 12th data driven control and learning systems conference (DDCLS)

作者： Wu, Hao Zhao, YingHao Yang, Xu Huang, Jian Cuil, Jiarui Univ Sci & Technol Beijing Sch Automat & Elect Engn Minist Educ Key Lab Knowledge Automat Ind Proc Beijing 100083 Peoples R China

ISBN: (纸本)9798350321050

driven by the increasing needs for production safety, a fault detection method based on multi-sensor fusion with adaptive weight coefficients is proposed in this paper to make full use of multi-measuring points information. To this end, considering the different information among multi-measuring points, the variance contribution rate (VCR) of vibration signals are used to design adaptive weight coefficients for data fusion to fully utilize the information contained in each vibration signal. On this basis, the least atoms contain time domain and frequency domain are extracted based on dictionary sparse representation (DSR) algorithm to represent the feature information of the original signal to weaken the influence of the curse of dimensionality. Finally, K-nearest neighbor distance is used in sparse residual space (SRS) for fault detection (K-SRS). The effectiveness of the proposed method is demonstrated by the rolling bearings data, and results show the advantage of our proposed approach.

关键词： multi-sensor information fusion variance contribution rate dictionary learning sparse decomposition sparse residual space

来源：评论

学校读者我要写书评

暂无评论

data-driven Distributed learning control for High-Speed Trains Considering Quantization Effects and Measurement Bias

引用

ieee TRANSACTIONS ON VEHICULAR TECHNOLOGY 2024年第7期73卷 9645-9655页

作者： Huang, Deqing Yu, Wei Shen, Dong Li, Xuefang Southwest Jiaotong Univ Sch Elect Engn Chengdu 610031 Peoples R China Southwest Jiaotong Univ Sch Elect Engn Chengdu 610031 Peoples R China Renmin Univ China Sch Math Beijing 100872 Peoples R China Sun Yat Sen Univ Sch Intelligent Syst Engn Guangzhou 510275 Peoples R China

The advanced train-to-train (T2T) communication technology, equipped with multiple high-speed trains (MHSTs), has the potential to enable train groups to maintain a stable T2T distance and achieve consensus tracking of MHSTs, thereby enhancing operational safety and efficiency. This study focuses on the data-driven distributed control issue of MHSTs considering quantization effects and measurement bias, employing a learning approach. Firstly, an equivalent linearization model of MHSTs and a transmission model accounting for sensor bias are constructed. Subsequently, a distributed model free adaptive iterative learning control (MFAILC) scheme using quantized signals is proposed. We then prove that the tracking error under the quantizer-based MFAILC is uniformly ultimately bounded, followed by further investigation on the impact of uniform quantizers. Finally, through a series of test conducted on the StarSim hardware-in-loop (HIL) semi-physical platform using quantified indicators, both the learning advantages of MFAILC and the influence of the quantization mechanism and measurement bias on MHSTs are verified.

关键词： Quantization (signal) Adaptation models Multi-agent systems Communication networks Analytical models Actuators Topology data-driven control distributed model free adaptive iterative learning control (MFAILC) multiple high-speed trains (MHSTs) quantization effects measurement bias

来源：评论

学校读者我要写书评

暂无评论

Dissipative Consensus via ILC of Singular Multiagent systems 13

Dissipative Consensus via ILC of Singular Multiagent Systems

引用

13th ieee data driven control and learning systems conference, DDCLS 2024

作者： Zhang, Meiyu Tian, Senping Gu, Panpan Li, Xiangyang School of Automation Science and Engineering South China University of Technology Guangzhou510641 China School of Electrical Engineering and Automation Hefei University of Technology Hefei230009 China

ISBN: (纸本)9798350361674

In this paper, the problem of dissipative consensus iterative learning control (ILC) is studied for singular multiagent systems (MASs). Firstly, a novel ILC algorithm is designed for such singular MASs. Then, under a connected communication graph, a sufficient condition is presented to make dissipative singular MASs abtain precise consensus tracking within a certain time interval. Finally, a simulation is conducted to validate the effectiveness of the proposed method. © 2024 ieee.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Reinforcement learning-based data-driven control Design for Motion control systems 36

Reinforcement Learning-based Data-driven Control Design for ...

引用

36th Chinese control and Decision conference (CCDC)

作者： Deng, Zhengqi Huo, Xin Du, Qinlong Liu, Qingquan Harbin Inst Technol Control & Simulat Ctr Harbin 150080 Peoples R China

ISBN: (纸本)9798350387780;9798350387797

Motion control systems are widely used in many fields of industry. Conventional control schemes are highly dependent on the system model to be designed. The performance of design would be greatly reduced, when the system exists unknown disturbances or uncertainty. Therefore, some scholars pointed out that the dependency on the system models can be eliminated by data-driven design schemes. In this paper, the reinforcement learning-based methods are included, which appeal to attentions gradually. The disturbances rejection problem for motion control systems is studied based on reinforcement learning. Considering the continuity of state space and action space, a method based on deep reinforcement learning algorithm is proposed to reject the periodic disturbances. Proposed deep deterministic policy gradient (DDPG) and twin delayed deep deterministic policy gradient (TD3) based algorithms are compared in simulation. The simulation results show that the periodic disturbances of the motion control systems can be rejected effectively with the proposed reinforcement learning controller.

关键词： Motion control systems data-driven control Reinforcement learning (RL) Deep deterministic policy gradient (DDPG) Twin delayed deep deterministic policy gradient (TD3)

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：